Many many company websites and reviews can give you information regarding this.

www.icb.uncu.edu.ar/upload/dnasequencing.pdf

http://www.oxbridgebiotech.com/review/research-and-policy/whats-so-special-about-next-generation-sequencing/

http://www.nature.com/nrg/journal/v11/n1/full/nrg2626.html

In principle Sanger sequencing requires a pure DNA sample (whether cloned or PCR amplified) with a possible primer binding site. the primer should be amplifying a specific product which gives the sequence, so very often cloned DNA / gel purified PCR products are necessary.

The Pyro/deep/next generation sequencing approaches usually use short DNA sequence mixtures, ligate adapters and consider each DNA fragment separately for sequencing. This makes sequencing of a mixture of DNA possible, avoiding cloning, only limitation being the read length.

Many many company websites and reviews can give you information regarding this.

www.icb.uncu.edu.ar/upload/dnasequencing.pdf

http://www.oxbridgebiotech.com/review/research-and-policy/whats-so-special-about-next-generation-sequencing/

http://www.nature.com/nrg/journal/v11/n1/full/nrg2626.html

In principle Sanger sequencing requires a pure DNA sample (whether cloned or PCR amplified) with a possible primer binding site. the primer should be amplifying a specific product which gives the sequence, so very often cloned DNA / gel purified PCR products are necessary.

The Pyro/deep/next generation sequencing approaches usually use short DNA sequence mixtures, ligate adapters and consider each DNA fragment separately for sequencing. This makes sequencing of a mixture of DNA possible, avoiding cloning, only limitation being the read length.

In sanger sequencing, you get a mean readout from the chromatogram peaks for each base. You usually call mixes when the DNA product harbours different bases at same position. In pyrosequencing, and other NGS technologies, you are able to sequence different DNA template separately in such a way that each sequence readout is linked to an initial single molecule DNA template which is amplified separately from other templates.

One more thing, with pyroseq you get very good reads from the first nucleotide on until 80 or max 200 bases, with sanger the first 20-40 bases are of poor quality but  reads upon 1kb are possible. 

@Naji

I think you are talking about the Roche 454 (pyro)sequencing. I was talking about the pyromark streptavidin coupled pyrosequencing.

Here is a paper comparing these two methods:

http://www.biomedcentral.com/1472-6750/11/6

cheers

In both cases you need cloning. In Sanger method you need to isolate one clone. In massive sequencing you can sequence let say all clones you have amplified. Sanger method is a method of choice if you like to sequence single clones you have selected or constructs you have made.

Let's refer to the basic 1-well-1-sequence scheme by instruments distributed originally by Biotage, now taken over by Qiagen, since NGS applications bring an overload of additional technical aspects.

An interesting part of the question is "Why DNA cloning is not necessary using pyrosequencing approach ?" In fact, from the very beginning, pyrosequencing was realized on in-vitro-amplified DNA, for at least two reasons:

(1) Like Sanger sequencing, it relies on a polymerase reaction that is primed by a DNA primer. Since the reaction takes place at comp. low temperature (37 deg). To ensure specific primer annealing, the template molecule should be as short as possible.

(2) The sequencing template must be single-stranded DNA to ensure primer annealing and efficient nt incorporation by polymerase. The recommended procedure will incorporate a biotin tag to the desired DNA strand (via PCR), DNA strand separation by alkaline buffer, binding of the target strand to streptavidin magnetic beads and washing.

Given these in-vitro preparation steps it was natural to use PCR for template amplification. PCR enables introduction of the biotin tag, and the intrinsic amplification makes time-consuming bacterial cloning unnecessary.

Important to say that in-vitro amplification and bacterial cloning do produce template samples with different characteristics. In-vitro amplification works for all template molecules that have proper primer binding sites. A template mixture will be reproduced quantitatively, although length differences may result in selective bias during PCR. As mentioned by the colleagues, this is the basis for applications like SNP typing, methylation analysis or (to add this) quantification of mRNA splicing variants:

http://dx.doi.org/10.1002/elps.200900292

If you sequence a mixture of DNA using Sanger, you will not get a clear signal, because you only get one read that gives you the "average" of all sequences in the mix. This is why you need to isolate and amplify clones before Sanger : To get one clean read per clone.

When using pyrosequencing (or any other NGS), you will get many reads. Each read corresponds to a single DNA molecule in the mix. So if you have an equal mix of 10 different DNAs, you should get roughly 10% of reads for each DNA. In this sense, you do not have to clone before pyrosequencing to get clean reads and know what your sample is compsed of. However, if you want to continue your work with one specific clone, this information will not help you: you will need to isolate and amplify one clone with your desired sequence, and in this case, Sanger is the method of choice.

pyrosequencing differs from sangers sequencing in that it relies on the detection of pyrophoshate release on nucleotide incorporation rather than chain termination with dideoxynucleotides.

Sanger sequencing generates long read lengths which are of not really good quality. This sequencing method requires good quality DNA resulting from good PCR and cloning. 

Pyrosequencing is categorised as NGS technology generating massive short reads per reaction. The reads are then assembled together giving a wider coverage and accuracy on the sequences.